An End-to-End Decision-Aware Multi-Scale Attention-Based Model for Explainable Autonomous Driving

2026-04-30

AI Summary

Current deep learning models for autonomous driving lack reliable and interpretable decision-making mechanisms, making it difficult to assess the reasonableness of their behavior and overall system reliability. This work proposes a decision-aware, multi-scale attention-based end-to-end model that simultaneously generates driving decisions and produces contextual explanations. The study introduces a novel Joint F1 score metric to more accurately evaluate model interpretability. Experimental results demonstrate that the proposed method significantly outperforms both classical and state-of-the-art approaches on the BDD-OIA and nu-AR datasets, exhibiting superior generalization capability and robustness.

Abstract

The application of computer vision is gradually increasing across various domains. They employ deep learning models with a black-box nature. Without the ability to explain the behavior of neural networks, especially their decision-making processes, it is not possible to recognize their efficiency, predict system failures, or effectively implement them in real-world applications. Due to the inevitable use of deep learning in fully automated driving systems, many methods have been proposed to explain their behavior; however, they suffer from flawed reasoning and unreliable metrics, which have prevented a comprehensive understanding of complex models in autonomous vehicles and hindered the development of truly reliable systems. In this study, we propose a multi-scale attention-based model in which driving decisions are fed into the reasoning component to provide case-specific explanations for each decision simultaneously. For quantitative evaluation of our model's performance, we employ the F1-score metric, and also proposed a new metric called the Joint F1 score to demonstrate the accurate and reliable performance of the model in terms of Explainable Artificial Intelligence (XAI). In addition to the BDD-OIA dataset, the nu-AR dataset is utilized to further validate the generalization capability and robustness of the proposed network. The results demonstrate the superiority of our reasoning network over the classic and state-of-the-art models.

Problem

• Explainable Artificial Intelligence
• Autonomous Driving
• Decision-Aware
• Model Interpretability
• Black-Box Models

Innovation

• decision-aware
• multi-scale attention
• explainable AI
• Joint F1 score
• autonomous driving